Federal Deposit Insurance Corporation as Receiver for Amcore Bank NA & Ors v Barclays Bank Plc & Ors

Disclosure sought to support expert evidence must satisfy the court, not merely the expert, that it is relevant, necessary and proportionate. An expert’s preference for additional data, or assertion that it would improve robustness, is insufficient where the expert has not analysed substantial data already disclosed. The court may assess necessity critically without deciding which competing expert methodology is correct. Disclosure of less direct comparators is not required merely because direct comparator data contains gaps, particularly where the pleaded case concerns systemic conduct capable of being inferred from available data. Wider disclosure may also be refused where it would disrupt the trial timetable.

The claimant, receiver for 19 closed US banks, alleged that the defendant banks colluded to submit artificially low USD LIBOR rates between August 2007 and December 2009. At the sixth case management conference, the claimant sought further transaction data for use by its expert in a proposed regression analysis of the alleged suppression.

The defendants had already disclosed millions of directly relevant borrowing and lending transactions covering the alleged suppression period and substantial periods before and after it. The claimant’s expert accepted that an analysis could be conducted using the existing data, but maintained that further data would make it more robust. The central issue was whether the additional disclosure was appropriate, relevant, necessary and proportionate.

1. Application dismissed. The application for further transaction data was refused, save for disclosure which NatWest and UBS had already agreed to provide.
2. The court, rather than the expert, determines whether disclosure is relevant, necessary and proportionate. The fact that an expert would prefer additional material or considers it ideal does not establish that disclosure is required.
3. The claimant had possessed extensive transaction data for more than three years but had not conducted the proposed regression analysis. Its expert accepted that the analysis could be performed using the existing data. In those circumstances, the court could not be satisfied that the additional data was necessary or required.
4. The alleged gaps were overstated by excluding non-London and non-interbank transactions, which remained more direct comparators with USD LIBOR than floating-rate notes and repos. The claimant’s case concerned sustained systemic suppression. If available data demonstrated a persistent pattern, the absence of data for particular days would not necessarily undermine the inference that the pattern continued.
5. The existing disclosure provided ample clean periods before and after the pleaded suppression period. Data further removed in time was likely to be less useful because the market and LIBOR submission processes changed. Additional data concerning secured or otherwise dissimilar transactions would also generate satellite technical disputes.
6. Ordering the disclosure would materially disrupt the timetable for expert reports and make adjournment of the 19-week trial likely. The disclosure was therefore disproportionate in addition to being unnecessary.